import streamlit as st # type: ignore
from streamlit_option_menu import option_menu # type: ignore
import streamlit_shadcn_ui as ui # type: ignore
from streamlit_echarts import st_echarts
import numpy as np
import pandas as pd
import seaborn as sns # type: ignore
import matplotlib.pyplot as plt
import folium #type:ignore
from streamlit_folium import st_folium #type:ignore
import plotly.express as px
import base64
import pickle
import time
from datetime import datetime
from pycaret.regression import load_model, predict_model


st.set_page_config(
    page_title="WALPA - Walmart Prediction App",
    page_icon="🧊",
    layout="wide",
    initial_sidebar_state="expanded",
)
@st.cache_data
def load_data(dataset):
    df = pd.read_csv(dataset)
    return df
def csvdownload(df):
    csv = df.to_csv(index=False)
    b64 = base64.b64encode(csv.encode()).decode()  # strings <-> bytes conversions
    href = f'<a href="data:file/csv;base64,{b64}" download="{df}_prediction.csv">Download CSV File</a>'
    return href
def autoplay_audio(file_path: str):
    with open(file_path, "rb") as f:
        data = f.read()
        b64 = base64.b64encode(data).decode()
        md = f"""
            <audio controls autoplay="true">
            <source src="data:audio/wav;base64,{b64}" type="audio/wav">
            </audio>
        """
        st.markdown(
            md,
            unsafe_allow_html=True,
        )

data = load_data('./datasets/Walmart.csv')
sumSales = data['Daily_Sales'].sum()
sumUnem = data['Unemployment'].sum()
def sum_Sales():
    if sumSales > 999 and sumSales < 9999:
        sum_display = "$" + str(sumSales)[:1] + "K" 
    elif sumSales > 9999 and sumSales < 99999:
        sum_display = "$" + str(sumSales)[:2] + "K"
    elif sumSales > 99999 and sumSales < 999999:
        sum_display = "$" + str(sumSales)[:3] + "K"
    elif sumSales > 999999 and sumSales < 9999999:
        sum_display = "$" + str(sumSales)[:1] + "M"
    elif sumSales > 9999999 and sumSales < 99999999:
        sum_display = "$" + str(sumSales)[:2] + "M"
    elif sumSales > 99999999 and sumSales < 999999999:
        sum_display = "$" + str(sumSales)[:3] + "M"
    elif sumSales > 999999999 and sumSales < 9999999999:
        sum_display = "$" + str(sumSales)[:1] + "MD"
    elif sumSales > 9999999999 and sumSales < 99999999999:
        sum_display = "$" + str(sumSales)[:2] + "MD"
    elif sumSales > 99999999999 and sumSales < 99999999999:
        sum_display = "$" + str(sumSales)[:3] + "MD"
    elif sumSales > 999999999999 and sumSales < 999999999999:
        sum_display = "$" + str(sumSales)[:4] + "MD"
    return sum_display

def sumUnemp():
    if sumUnem > 999 and sumUnem < 9999:
        sum_Unem = str(sumUnem)[:1] + "K" 
    elif sumUnem > 9999 and sumUnem < 99999:
        sum_Unem = str(sumUnem)[:2] + "K"
    elif sumUnem > 99999 and sumUnem < 999999:
        sum_Unem = str(sumUnem)[:3] + "K"
    elif sumUnem > 999999 and sumUnem < 9999999:
        sum_Unem = str(sumUnem)[:1] + "M"
    elif sumUnem > 9999999 and sumUnem < 99999999:
        sum_Unem = str(sumUnem)[:2] + "M"
    elif sumUnem > 99999999 and sumUnem < 999999999:
        sum_Unem = str(sumUnem)[:3] + "M"
    elif sumUnem > 999999999 and sumUnem < 9999999999:
        sum_Unem = str(sumUnem)[:1] + "MD"
    elif sumUnem > 9999999999 and sumUnem < 99999999999:
        sum_Unem = str(sumUnem)[:2] + "MD"
    elif sumUnem > 99999999999 and sumUnem < 99999999999:
        sum_Unem = str(sumUnem)[:3] + "MD"
    elif sumUnem > 999999999999 and sumUnem < 999999999999:
        sum_Unem = str(sumUnem)[:4] + "MD"
    return sum_Unem

def main():
    with st.sidebar:
        
        selected = option_menu("Main Menu", ['Home', 'Dashboard', 'Analysis', 'Visualization', 'Machine Learning'], 
            icons=['house','speedometer2', 'boxes', 'graph-up-arrow', 'easel2'], menu_icon="list", default_index=0,
            styles={
                "container": {"padding": "5px", "background-color": "transparent", "font-weight": "bold"},
                "icon": {"font-size": "17px"}, 
                "nav-link": {"font-size": "15px", "text-align": "left", "margin":"5px","padding": "10px", "--hover-color": "#1E90FF"},
                "nav-link-selected": {"background-color": "#1E90FF"},
            }
        )
    # Subdivide the page into three columns
    left,middle,right = st.columns((0.5,4,0.5))
    if selected == 'Home':
        with middle:
            col1, col2, col3 = st.columns(3)
            with col2:
                st.image('./assets/images/walpa-logo.png')
            st.subheader('What is Walpa ?')
            st.write("Walpa is a Streamlit Machine Learning App created to assist data engineers in multiple tasks such as datasets Analysis report, visualization, and predictions for the case of Walmart Inc.")
            st.write("This is not an official Walmart Inc app is just for educational purpose")
            st.subheader("Walpa's Team")
            team = [
                {"role": "Founder", "name": "Jason Ntone"},
                {"role": "Developer", "name": "Jason Ntone"},
                {"role": "Designer", "name": "Jason Ntone"}
            ]
            st.write(team)
            st.markdown(" - All rights reserved WALPA\u00A9")
    elif selected == 'Dashboard':
        # First row
        with middle:
            col1, col2, col3 = st.columns(3)
            with col2:
                st.image('./assets/images/walpa-logo.png')
            st.title("Walmart Dashboard")
            col4, col5, col6 = st.columns(3)
            with col4:
                temp = st.metric(label="Total Sales", value=sum_Sales(), delta="From 5010 To 2012")
            with col5:
                temp = st.metric(label="Total Unemployemt", value=sumUnemp(), delta="From 2010 To 2012")
            with col6:
                temp = st.metric(label="Total Stores studied", value=45, delta="From 2010 To 2012")

        with middle:
            st.subheader("Walmart Stores Map")

            stores = data['Store'].unique()
            longitude_values = [-111.0327, -88.1668, -121.3477, -77.0891, -87.3695, -95.3271, -79.2854, -84.3594, -81.5951, -82.7852, -118.5694, -82.2711, -80.6665, -78.2971, -103.3284, -84.8482, -93.0727, -117.0266, -97.0088, -82.1349, -76.8572, -104.7973, -123.2838, -91.5127, -117.3879, -97.9895, -80.2403, -82.0174, -94.6041, -117.0774, -88.2285, -81.4383, -83.3702, -93.2422, -100.4930, -81.8765, -85.4835, -117.0731, -79.7245, -86.2356, -75.7216, -90.1516, -77.8990, -86.2169, -96.6857]
            latitude_values = [32.1555, 39.4931, 37.9886, 38.7684, 36.5298, 29.5636, 33.3776, 33.7603, 31.8469, 39.9673, 34.2801, 27.9944, 37.1505, 36.0659, 34.1866, 37.8041, 44.8955, 32.9759, 30.6631, 33.5412, 39.6366, 41.1364, 44.5714, 31.5634, 34.1041, 26.1536, 39.0212, 38.9188, 38.8837, 32.6389, 42.9937, 30.2862, 33.3263, 45.1571, 28.7043, 27.2008, 39.3378, 32.6072, 39.9002, 32.3838, 40.8332, 32.4081, 34.1641, 32.3418, 40.7399]
            
            # Create a map
            wmap = folium.Map(location=[37.0902, -95.7129], zoom_start=4)

            # Add markers for each store
            for store, lon, lat in zip(stores, longitude_values, latitude_values):
                folium.Marker([lat, lon], popup=store,icon=folium.Icon(color='blue', icon='shopping-cart', prefix='fa')).add_to(wmap)

            # Fit the map to the bounds of the USA
            wmap.fit_bounds([[24.396308, -125.000000], [49.384358, -66.934570]])
            # call to render Folium map in Streamlit
            st_data = st_folium(wmap, width=800)
    elif selected == 'Analysis':
        with middle:
            col1,col2,col3 = st.columns((0.5,3,0.5))
            with col2:
                tab = ui.tabs(options=['Overview', 'Sumary', 'Correlation Matrix'], default_value='Overview', key="none")
            st.title("Data Analysis")
            if tab == 'Overview':
                st.subheader('Walmart Daily Sales Overview')
                st.dataframe(data.head())
            elif tab == 'Sumary':
                st.subheader('Walmart Daily Sales Sumary')
                st.dataframe(data.describe())
            elif tab == 'Correlation Matrix':
                st.subheader('Walmart Correlation Matrix')
                fig = plt.figure(figsize=(15,5))
                st.write(sns.heatmap(data.corr(),annot=True))
                st.pyplot(fig)
    elif selected == 'Visualization':
        with middle:
            tab = ui.tabs(options=['Regplot', 'Barplot', 'Lineplot'], default_value='Barplot', key="none")
            if tab == 'Regplot':
                st.subheader('Walmart Daily Sales Regplot')
                fig = plt.figure(figsize=(15,5))
                st.write(sns.regplot(data=data, x='Store', y='Daily_Sales'))
                st.pyplot(fig)
            elif tab == 'Barplot':
                st.subheader('Walmart Daily Sales Barplot')
                option = {
                    "xAxis": {
                        "type": "category",
                        "data": data['Store'].tolist(),
                    },
                    "yAxis": {
                        "type": "value"
                    },
                    "series": [{
                        "data": data['Daily_Sales'].tolist(),  # Replace 'Sales' with the actual column name for sales data
                        "type": "bar"
                    }]
                }
                st_echarts(
                    options=option,
                    height="400px",
                )
            elif tab == 'Lineplot':
                st.subheader('Walmart Daily Sales line plot')
                option = {
                    "xAxis": {
                        "type": "category",
                        "data": data['Date'].tolist(),
                    },
                    "yAxis": {
                        "type": "value"
                    },
                    "series": [{
                        "data": data['Daily_Sales'].tolist(),  # Replace 'Sales' with the actual column name for sales data
                        "type": "line"
                    }]
                }
                st_echarts(
                    options=option,
                    height="400px",
                )

    elif selected == 'Machine Learning':
        with middle:
            st.subheader('📈🎯 Daily Sales Prediction Widget')
            tab = ui.tabs(options=['Method 1: Upload Dataset', 'Method 2: Fill the form'], default_value='Fill the form', key="none")
            st.write('\n')
            if tab == 'Method 2: Fill the form':
                st.markdown('**Fill the form with correct data to make prediction**')
                col1,col2,col3 = st.columns(3)
                with col1:
                        
                    st.write('Enter the Year')
                    year = ui.input(type='number', default_value=0, key="year")
                with col2:
                    st.write('Enter the Month')
                    month = ui.input(type='number', default_value=0, key="month")
                with col3:
                    st.write('Enter the Day')
                    day = ui.input(type='number', default_value=0, key="day")
                    
                        
                st.write('Enter Store Number')
                store = ui.input(type='number', default_value=0, key="input1")
                    
                st.write(f'The date : **{year}-{month}-{day}** you have entered is that a holiday ?')
                holiday = [
                    {"label": "Yes", "value": 1, "id": "r1"},
                    {"label": "No", "value": 0, "id": "r2"},
                ]
                holiday_flag = ui.radio_group(options=holiday, default_value=0, key="radio1")
                
                col3,col4 = st.columns((2,2))
                with col3:
                    st.write('Enter the Temperature')
                    temperature = ui.input(type='text', default_value=0, key="input2")
                        
                with col4:
                    st.write('Enter the Fuel Price')
                    fuel_price = ui.input(type='text', default_value=0, key="input3")
                        
                col5,col6 = st.columns((2,2))
                with col5:
                    st.write('Enter the CPI')
                    cpi = ui.input(type='text', default_value=0, key="input4")
                        
                with col6:
                    st.write('Enter the Unemployment')
                    unemployment = ui.input(type='text', default_value=0, key="input5")
                
                    Store= int(store)
                    Holiday_Flag= int(holiday_flag)
                    Temperature= float(temperature)
                    Fuel_Price= float(fuel_price)
                    CPI= float(cpi)
                    Unemployment= float(unemployment)
                    Year = int(year)
                    Month = int(month)
                    Day = int(day)
                    form_data = pd.DataFrame([[Store, Holiday_Flag, Temperature, Fuel_Price, CPI, Unemployment, Year, Month, Day]],
                        columns=['Store', 'Holiday_Flag', 'Temperature',
                            'Fuel_Price', 'CPI', 'Unemployment','Year', 'Month', 'Day'])
                    st.subheader('Your provided data')
                    st.dataframe(form_data)
                    submit_btn = ui.button(text="Predict Daily Sales", key="styled_btn_tailwind", className="bg-blue-500 text-white")
                    if submit_btn:
                        if form_data.empty == False:
                            model_path = './models/Walmart'
                            model = load_model(model_path)

                            # Make predictions using the loaded model and form_data
                            pred = predict_model(model, data=form_data)

                            prediction = pred['prediction_label'].values[0]

                            # Display the prediction
                            with st.status("Daily Sales Prediction processing...", expanded=True) as status:
                                st.write("Handling data...")
                                time.sleep(2)
                                st.write("Load Model...")
                                time.sleep(1)
                                st.write("Load Data...")
                                time.sleep(1)
                                status.update(label="Daily Sales Prediction processing complete!", state="complete", expanded=False)
                                autoplay_audio("./assets/audio/mixkit-positive-notification-951.wav")
                            st.success(prediction)
                        else:
                            st.warning("Fill the form")
            if tab == 'Method 1: Upload Dataset':
                with middle:
                    st.header("Load your file")
                    uploaded_file = st.file_uploader('Upload your Dataset(.csv file)',
                        type=['csv'])
                    if uploaded_file:
                        df = load_data(uploaded_file)
                        df['Year'] = pd.to_datetime(df['Date']).dt.year
                        df['Month'] = pd.to_datetime(df['Date']).dt.month
                        df['Day'] = pd.to_datetime(df['Date']).dt.day
                        df = df.drop(['Date'], axis=1)
                        model_path = './models/Walmart'
                        model = load_model(model_path)
                        preds = predict_model(model, data=df)
                        # Assuming 'predictions' is a list or array-like object
                        predictions = preds['prediction_label'].values

                        # Create DataFrame
                        pp = pd.DataFrame(predictions, columns=['Daily_Sales_Prediction'])

                        ndf = pd.concat([df, pp], axis=1)
                        st.subheader("Daily_Sales Predictions")
                        with st.status("Daily Sales Prediction processing...", expanded=True) as status:
                            st.write("Handling data...")
                            time.sleep(2)
                            st.write("Load Model...")
                            time.sleep(1)
                            st.write("Load Data...")
                            time.sleep(1)
                            status.update(label="Daily Sales Prediction processing complete!", state="complete", expanded=False)
                            autoplay_audio("./assets/audio/mixkit-positive-notification-951.wav")
                        st.write(ndf)  

                
if __name__ == '__main__':
    main()